Swapnil: Can you tell us a bit about yourself? How you got associated with computers?
Stefano: With computers … wow, that’s quite some time ago! I discovered programming when I was 9 year old – I’m 31 now – finding an illustrated book about BASIC programming at the local library (!). Pair that with a C64 that I had at home and a young geek was born.
Unfortunately, I discovered GNU/Linux only many many years later, in 1998, when as a 2nd year computer science student I attended an operating system course taught by a Free Software “evangelist” (which I will never thank enough for that).
Nowadays in addition to my Debian involvement, which I do as a volunteer, I’m immersed in Free Software also as part of my day job at IRILL (http://www.irill.org), a newly founded computer science research center on Free Software related topics.
Swapnil: How does Free Software matter for ordinary people? Should they care, the way they care for organic food, CO2 emission and other sustainable methods? Can we compare Free Software with such causes?
Stefano: I think Free Software matters to “ordinary people” and that it will soon be as relevant for their lives as the other causes you mention already are. Software (be it Free or not) is more and more pervasive in our lives. It is software that intervenes when we drive a car, it is software that handles our private data within social networks, and it is software that controls the medical devices which cure and keep us alive. Tasks that were once delegated to mechanical devices are more and more delegated to software. It is simply unthinkable to lose the freedoms we once had with mechanical devices – such as bringing them to some expert friend for inspection, repair, or modification – only because computers have become popular.
Free Software is the only defense people have to retain those dear old freedoms. Without Free Software, people will stop owning the devices that run their lives.
Swapnil: What is your preferred DE and why?
Stefano: I use the GNOME Desktop Environment with Awesome as a window manager.
GNOME because it is a very polished and well-integrated desktop environment, which doesn’t get too much in my way. Awesome, instead of the default GNOME window manager, because I want a tiling window manager with a mature implementation.
Swapnil: Debian has hit the version 6 this month. How does it feel like being the project leader of the most reputed Linux distribution?
Stefano: It’s hard to describe actually. It’s a honor to be the “face” of a project with such a history and such an important role in Free Software. But more than that, I guess the main feeling is that of happiness in having the possibility of *helping* the Debian community having fun in doing what they want to do: making the word a better place by creating a Free operating system. It’s quite a responsibility, but it totally pays back!
Swapnil: What drove the inclusion of BSD port?
Stefano: As it often happens in Debian, the main driving force has simply been the enthusiasm and skills of a group of geeks who wanted to make it true. The Debian GNU/kFreeBSD team deserves all the kudos for that. As it often happens for ports though, similar enthusiasms are driven by the desire of having Debian “conquer” yet another system. I confess that the idea of having the first non-Linux port released, even if only as a technology preview, has a lot of charm. With ports to new kernels, one also gets a payoff in terms of features supported by the target kernel (such as ZFS support in the case of Debian GNU/kFreeBSD) and they play a role in motivating geeks as well.
Swapnil: You have been re-elected as the Debian Leader, what are the new challenges and goal for you? Could you also mention some of your major achievements as the Debian leader?
Stefano: No, I could not mention my own achievements, because Debian is a team sport. I might have catalyzed some achievements, but I’m far from being entitled to claim them as achievements of mine. That said, there have been several Debian happenings during the past year that I’m particularly happy and proud of. One is undoubtedly the Squeeze release and some of its novelties, such as the removal of non-free firmware blobs and the kFreeBSD ports. Another one, at a more social level, is the recognition of full Debian project membership status (AKA “Debian Developer” status) to contributors with abilities other than packaging.
Monocultures are unhealthy in general and Free Software is no exception;taking that step Debian has made an important leap forward in bringing useful diversity into the Debian community.
Several challenges are ahead of us. A particularly important one for me is to keep on showing to the Free Software world that an independent, volunteer-based distro can compete with distros which are sponsored by individual companies. Independence is the only guarantee that money interest will not prevail over software freedom interests and Debian has an important role to play there. Debian is, after all, one of the very few independent distros among the “most popular distros” that cannot be pinned to a company.
That does not mean that Debian lives in an ideal world where companies do not have a role to play in Free software. Quite the contrary: one of our future challenges is indeed to find ways to encourage companies to contribute work to Debian, without letting that get into the way of existing Debian decision making processes and independence.
Another challenge ahead of us is how to actually put into use the diversity that, potentially, generalized project membership could bring.
Historically, in Debian we have always had troubles attracting people with non-packaging skills, such as graphic and media artists, documentation writers, journalists, etc. At the same time other communities as geek as Debian seem to be very good at attracting those contributions. I believe we are well past the vision that those skills are useless for a Free Software project (they are not), but we still need to find our way to attract non-packaging contributions.
Swapnil: Debian 6 is here with mixed reviews. Debian vanilla is still a bit hard for an average user to use. Is there any change in the approach to make Debian usable even for average Joe or will Debian remain a platform for distros like Ubuntu?
Stefano: Every time I read expressions as “average Joe” I shiver a bit. There is no such an average users, there are several different types of users. Some distros are targeted at mass market desktop users, which arguably are *a* specific kind of user, Debian simply is not. We try to be general and customizable enough for several different use cases and we do offer several profiles to address the needs of the kind of users we are aware of. We do that by the means of both task selections at the end of the installation and by the means of pre-customized Debian images known as Debian Pure Blends.
Focusing on desktop users (your “average Joe”, I believe) I think the usability gaps lay into two main camps. The first one is polishing: it’s undeniable that a Debian default desktop installation is not as eye-candy and polished as the offering of other desktop-oriented
distros. On that, we get back to our ability to attract contributions from graphic artists and the like. We need to get better at that, given everything else needed to be as polished as other desktop distros is already available.
The second camp is hardware support: we do fight the battle of Free Software and part of that battle is discriminating non-free software. At a times, that implies extra installation burden for the users. I think it’s the right approach nonetheless: it’s better to take that chance to explain to users why non-free software is bad for them, at the risk of losing some users, than pre-installing non-free software and further the (honest) ignorance of the risks connected to non-free software.
Swapnil: Could you tell us a bit about the organizational structure of Debian, what role the project leader has for Debian?
Stefano: Formally, the DPL (Debian Project Leader) is an official representative of the Debian Project both within the project and to the external world (conferences, trade show, “political” relationships with other actors of the Free Software ecosystem, etc.). The DPL is also a sort of “decision garbage collector” who is in charge of deciding on matters for which no specific Debian team exists, decides upon the usage of Project resources, and can delegate tasks to specific individuals or team.
Informally, the organizational structure of Debian is based on independent teams which are formed following a “do-ocratic” process.
Anyone who is willing to take over some task first needs to show they are able and willing to do it, and then naturally become part of the team who is care-taking at that task. In such an informal setting, the day to day job of the DPL is to shepherd the good health of the hundreds and hundreds of existing teams, as well as animate project-wide discussions on topics that escape the scope of a single team.
Swapnil: What is the status of Debian developers community? How do you plan to engage more green-horn developers to work on Debian?
Stefano: I don’t have numbers at hand, but the flow of new developers joining Debian is steady. Of course, there is also a flow of people leaving Debian (which often journalists pick on as a bad sign…), but in a Project as old as Debian (18 years and counting) such a turn-over is not only normal, but even healthy.
I think the best way to attract new developers is being clear about what Debian is and why it is different from other distributions. Debian is here to provide the best, entirely Free Software, operating system and wants to achieve that result by the means of a volunteer-based community united by shared ideals. If we manage to communicate that clearly, we will attract the right™ people we need to make Debian.
Swapnil: Are derivatives like Ubuntu, Mint affecting the dedicated Debian developer base, thinning it? From an opposite point of view, does Debian get any direct benefit from developers working on these derivatives?
Stefano: I don’t have any perception of Developers leaving Debian to join some of its derivatives, if that is what you mean. Rather, we get a fairly regular flow of new developers joining Debian coming from some of its derivatives. That is a very good sign and it means we are doing a good job of explaining that contributing to Debian (or, more generally, contributing “as much upstream as you can”) is the best way to contribute not only to your favorite distro, but also to Free Software in general.
Arguably the above is an indirect benefit and you were asking about direct benefits. In that respect we are catching up with initiatives like the Debian Derivatives FrontDesk and DEX (Debian dErivatives eXchange) that provide forums for and organize cross-distro initiatives that bring patches developed within derivatives back into Debian proper.
Of course it would be better to have work done directly into Debian from day 0 instead of having to catch up later on with such initiatives. But we do have a backlog of differences to work through and I do understand that different derivatives might have different constraints which are not always compatible with Debian constraints.
Swapnil: When we see derivatives like Ubuntu ‘branching’ off with stuff like Unity and Wayland, how does it affect Debian? What is your opinion about Unity and Wayland?
Stefano: From the Debian point of view, stuff like Unity and Wayland is upstream software, we can package it in our archive once we believe it meets our quality and licensing criteria. The reason there have been big fights around it in other distros is that they are distros with a specific user profile in mind (e.g. desktop user) and where defaults matter a lot. Changing those defaults can make the fortune (or the misery) of any given upstream software project, at least in terms of the size of its user base. In Debian this is less of a problem, as we are striving to be universal, with no single use case in mind. That comes with its own set of problems of course… and doesn’t relieve us of our own lovely debates over defaults!
Regarding Unity, I think that the general principle of experimenting with UIs is a good one and I salute whoever tries to innovate in that area. Nonetheless, what worries me about Unity is the grudges it has brought among the distribution editor (Ubuntu) and the upstream community (GNOME).
In Debian we know how much it’s important to have good relationships with upstreams. In particular, if you believe that the greater goal is the success of Free Software rather than the success of any single distro (as we do believe in Debian), you realize pretty quickly that working well with upstream is key. Undermining the relationships among the GNOME community and one of the biggest GNOME vendor brings very little benefits to the Free Software cause.
Swapnil: Debian support most architecture. Is Debian contemplating optimized versions for tablets and smartphones?
Stefano: Debian has had an ARM port well before it was “cool” to have one. During the past year we have organized a lot of sprints for Debian teams working on ARM ports, collaborating fruitfully with other actors such as Linaro and ARM itself. A Debian armhf (hard float) port is already in the working to support modern ARM boards; I’m confident it will be up to the quality requirements needed to be part of the Debian Wheezy release.
Swapnil: What is your opinion about Mono, considering Microsoft’s consistent attacks on Free Software? Is usage of Mono a threat for Free Software community?
Stefano: The real threat are software patents, more than any individual technology. As long as software patents will be around, traps will be everywhere, even in the most unexpected applications. This is so “thanks” to the little scrutiny to prior art patent offices around the world have made, a practice which is very unlikely to change that whose damage will stay with us as long as software patents exist.